The prosurvival Bcl-2 Homolog Bfl-1/A1 is a direct transcriptional target of NF-kB that blocks TNF-?? induced apoptosis

Abstract

Bcl-2-family proteins are key regulators of the apoptotic response. Here, we demonstrate that the pro-survival Bcl-2 homolog Bfl-1/A1 is a direct transcriptional target of NF-kappaB. We show that bfl-1 gene expression is dependent on NF-kappaB activity and that it can substitute for NF-kappaB to suppress TNFalpha-induced apoptosis. bfl-1 promoter analysis identified an NF-kappaB site responsible for its Rel/NF-kappaB-dependent induction. The expression of bfl-1 in immune tissues supports the protective role of NF-kappaB in the immune system. The activation of Bfl-1 may be the means by which NF-kappaB functions in oncogenesis and promotes cell resistance to anti-cancer therapy.

Full text

RESEARCH COMMUNICATION
The prosurvival Bcl-2 homolog
Bfl-1/A1 is a direct
transcriptional target of NF-␬B
that blocks TNF␣-induced
apoptosis
Wei-Xing Zong,
1,2
Leonard C. Edelstein,
1,2
Cailin Chen,
1
Judy Bash,
1,3,5
and Ce´line Ge´linas
1,4,6
1
Center for Advanced Biotechnology and Medicine,
2
Graduate
Program in Biochemistry and Molecular Biology, University
of Medicine and Dentistry of New Jersey–Robert Wood
Johnson Medical School (UMDNJ–RWJMS),
3
Graduate
Program in Microbiology and Molecular Genetics, Rutgers
University,
4
Department of Biochemistry, UMDNJ–RWJMS,
Piscataway New Jersey 08854-5638 USA
Bcl-2-family proteins are key regulators of the apoptotic
response. Here, we demonstrate that the pro-survival
Bcl-2 homolog Bfl-1/A1 is a direct transcriptional target
of NF-␬B. We show that bfl-1 gene expression is depen-
dent on NF-␬B activity and that it can substitute for
NF-␬B to suppress TNF␣-induced apoptosis. bfl-1 pro-
moter analysis identified an NF-␬B site responsible for
its Rel/NF-␬B-dependent induction. The expression of
bfl-1 in immune tissues supports the protective role of
NF-␬B in the immune system. The activation of Bfl-1
may be the means by which NF-␬B functions in onco-
genesis and promotes cell resistance to anti-cancer
therapy.
Received October 15, 1998; revised version accepted January
6, 1999.
Binding of the proinflammatory cytokine tumor necrosis
factor ␣ (TNF␣) to its receptor triggers competing signal-
ing pathways that determine whether a cell lives or dies.
Whereas one pathway is conducive to cell death, the
other leads to activation of Rel/NF-␬B transcription fac-
tors and the coincident inhibition of apoptosis (for re-
view, see Nagata 1997). Accumulating evidence supports
a proactive role for NF-␬B in the inhibition of cell death
induced by TNF␣ and other death-causing agents (for
review, see Van Antwerp et al. 1998). Whereas the acti-
vation of NF-␬B blocks cell killing, its inhibition en-
hances the cytotoxicity of TNF␣ and promotes apoptosis
in various cell systems, demonstrating the need for NF-
␬B function for cell survival (Beg et al. 1995; Beg and
Baltimore 1996; White et al. 1995; Liu et al. 1996; Van
Antwerp et al. 1996; Wang et al. 1996;Wu et al. 1996; Cai
et al. 1997; Zong et al. 1997). The protective effect of
NF-␬B is dependent on RNA and protein synthesis, sug-
gesting that it regulates the expression of genes that con-
fer resistance to death-inducing signals (for review, see
Nagata, 1997). The finding that an intact transactivation
domain is required for Rel/NF-␬B factors to block cell
death agrees with this hypothesis (Zong et al. 1998).
Proteins of the Bcl-2 family play critical roles in deter-
mining cell fate in the apoptotic pathway. Although
some members antagonize cell death, others exhibit a
proapoptotic activity (for reviews, see Reed 1996; Adams
and Cory 1998). By subtractive cDNA cloning, we iden-
tified the prosurvival Bcl-2-homolog Bfl-1/A1 as a direct
transcriptional target of NF-␬B. We show that ectopi-
cally expressed Rel proteins and stimuli that activate
endogenous NF-␬B factors up-regulate bfl-1 gene expres-
sion and that this is inhibited by a dominant I␬B␣⌬N
transgene. Expression of Bfl-1 alone conferred resistance
to TNF␣ cytotoxicity, indicating that it can substitute
for NF-␬B to suppress apoptosis. bfl-1 promoter analysis
identified a consensus NF-␬B site responsible for its Rel-
dependent induction. Together, these results demon-
strate that NF-␬B directly activates Bfl-1/A1 to inhibit
programmed cell death. The preferential expression of
bfl-1 in immune tissues supports the protective role of
NF-␬B in the immune system.
Results and Discussion
We used a PCR-selected subtractive cDNA cloning ap-
proach to identify anti-apoptotic genes under Rel/NF-␬B
control. The HeLa-derived HtTA–CCR43 cell line,
which conditionally expresses c-Rel under the control of
a tetracycline-regulated promoter, is resistant to TNF␣-
induced cell death upon induction of c-Rel (Bash et al.
1997; Zong et al. 1998). mRNA from HtTA–CCR43 cells
conditionally expressing c-Rel was reverse transcribed
and subjected to subtraction and PCR amplification. A
subtracted cDNA fragment of ∼700 bp was found to be
identical to bfl-1, a member of the Bcl-2 family of apop-
tosis inhibitors. Bfl-1 was originally isolated from fetal
liver and from cytokine-treated endothelial cells (Choi et
al. 1995; Karsan et al. 1996) and shares 72% amino acid
identity with its mouse homolog A1 (Lin et al. 1993).
Northern blot analysis during a time course of c-Rel
induction confirmed the up-regulation of bfl-1 tran-
scripts in HtTA–CCR43 cells, with kinetics that paral-
leled those of c-rel and of the Rel/NF-␬B target gene i
␬
b
␣
(Fig. 1a). Induction of the transactivation-competent
RelA subunit of NF-␬B also led to a sharp increase in
bfl-1 mRNA levels in the tetracycline-regulated HtTA–
RelA cell line (Fig. 1b, lanes 4–6; Zong et al. 1998). In
contrast no expression was detected in response to the
p50/NF-␬B1 protein, which lacks a defined transcription
activation domain (Fig. 1b, lanes 1–3). Thus, bfl-1 gene
expression was specifically up-regulated upon ectopic
expression of transcriptionally active Rel/NF-␬B sub-
units.
[Key Words: Rel; NF-␬B; Bfl-1; A1; TNF␣; apoptosis]
5
Present address: Cancer Institute of New Jersey, New Brunswick, New
Jersey 08854 USA.
6
Corresponding author.
E-MAIL gelinas@mbcl.rutgers.edu; FAX (732) 235-5289.
382 GENES & DEVELOPMENT 13:382–387 © 1999 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/99 $5.00; www.genesdev.org

Next, we verified the ability of endogenous NF-␬B fac-
tors to activate bfl-1 gene expression in response to vari-
ous stimuli and in different cells. TNF␣ strongly induced
bfl-1 transcripts in human HT1080 fibrosarcoma cells
(Fig. 2a, lanes 1,2). bfl-1 mRNA levels were also strongly
elevated in human Jurkat T-cells stimulated with phor-
bol 12-myristate 13-acetate (PMA) plus ionomycin (lanes
3,4). Likewise, the treatment of mouse 70Z/3 pre-B cells
with bacterial lipopolysaccharides (LPS) promoted the
accumulation of bfl-1/a1 transcripts (lanes 5,6). Consis-
tent with the nuclear NF-␬B activity found in mouse
WEHI-231 B cells (Liou et al. 1994; Miyamoto et al.
1994), basal bfl-1/a1 expression was observed in these
cells (lanes 7,9,11). bfl-1/a1 mRNAs were further in-
duced by different NF-␬B-inducing agents, including
TNF␣, LPS, or PMA (lanes 8,10,12).
The activity of NF-␬B is controlled by its association
with I␬B factors in the cytoplasm. In contrast to wild-
type I␬B␣, which undergoes proteasome-mediated degra-
dation in response to stimuli, an amino-terminally de-
leted I␬B␣ transgene is resistant to signal-induced degra-
dation and acts as a constitutive repressor of NF-␬Bin
Jurkat–I␬B␣⌬N cells (Chu et al. 1997). We investigated
whether NF-␬B was necessary for the stimuli-induced
activation of bfl-1 by characterizing the effects of wild-
type I␬B␣ and a trans-dominant I␬B␣⌬N inhibitor on
bfl-1 gene expression. The levels of 28S rRNA and of
mRNA for the NF-␬B-regulated gene interleukin-8 (IL-8)
were monitored as controls. Similar to the induction of
bfl-1 in the parental Jurkat T-cell line (Fig. 2a, lanes 3,4),
bfl-1 transcripts were sharply elevated by PMA plus
ionomycin treatment of T cells expressing a wild-type
I␬B␣ transgene (Fig. 2b, lanes 1,2). In contrast, the induc-
tion of bfl-1 was reduced by 83% and that of IL-8 was
decreased by 65% in cells expressing I␬B␣⌬N (lanes 3,4).
This indicated that nuclear NF-␬B activity is important
for bfl-1 gene expression.
Similar to c-rel, relA, and their target i
␬
b
␣
, bfl-1 was
found to be abundantly expressed in discrete immune
tissues. A Northern blot survey showed high levels of
bfl-1 transcripts in human spleen, lymph nodes, periph-
eral blood leukocytes and bone marrow (Fig. 3, lanes
6,7,9,10). Little expression was seen in thymus, fetal
liver, ovary, and small intestine, whereas none was de-
tected in prostate, testis, or colon (lanes 1–5,8,11). These
results suggest a physiological role for bfl-1 in promoting
the survival of defined cell lineages in the immune sys-
tem.
In agreement with the anti-apoptotic activity of NF-␬B
toward TNF␣, the transient cotransfection of a CMV-
bfl-1 expression vector with a CMV–␤-galactosidase re-
porter plasmid significantly suppressed the TNF␣-in-
duced killing of human HeLa cells in the presence of the
protein synthesis inhibitor cycloheximide (CHX; Fig.
4a). Quantitation of cell survival showed that the tran-
sient expression of bfl-1 increased the viability of HeLa
cells 8.5-fold in comparison to cells transfected with the
control pCMV plasmid (Fig. 4b). Cell protection was also
observed in the human HtTA-1 and HT1080 cell lines
(Fig. 4b). Moreover, Bfl-1 also suppressed TNF␣-induced
apoptosis under conditions in which endogenous NF-␬B
activity was inhibited by a serine-to-alanine mutant of
Figure 1. bfl-1 gene expression is activated by c-Rel and RelA
but not by p50/NF-␬B1. (a) Expression of bfl-1 transcripts in
HtTA–CCR43 cells maintained in the presence (lane 1)orab-
sence of tetracycline for 12, 24, 36, or 48 hr to induce c-rel
expression (lanes 2–5). The blot was successively hybridized to
32
P-labeled probes for bfl-1, i
␬
b
␣
,c-rel, and gapdh.(b) bfl-1 gene
expression in HtTA–p50 and HtTA–RelA cells maintained in
the presence (lanes 1,4) or absence of tetracycline for 24 or 48 hr
to induce the expression of p50 or relA (lanes 2,3,5,6). The blot
was hybridized to bfl-1 and actin probes.
Figure 2. The expression of bfl-1 is dependent on endogenous
Rel/NF-␬B activity. (a) The activation of NF-␬B induces bfl-1
gene expression. Human HT1080 cells untreated or treated with
TNF␣ (lanes 1,2), human Jurkat T cells treated with DMSO as a
control (lane 3) or with PMA plus ionomycin (lane 4), mouse
70Z/3 pre-B cells untreated or treated with LPS (lanes 5,6),
mouse WEHI-231 B cells untreated or treated with TNF␣ (lanes
7,8), LPS (lanes 9,10), or PMA (lanes 11,12). Total RNA (20 µg)
was hybridized to bfl-1 and actin probes. (b) The induction of
bfl-1 gene expression is repressed by a dominant I␬B␣⌬N inhibi-
tor. Jurkat T cells expressing wild-type I␬B␣ or an I␬B␣⌬N trans-
gene were treated with DMSO alone (lanes 1,3)orwithPMA
plus ionomycin (lanes 2,4). Total RNA (20 µg) was hybridized to
32
P-labeled probes for bfl-1, il-8, and 28S rRNA.
Bfl-1/A1 is under Rel/NF-␬B control
GENES & DEVELOPMENT 383

I␬B␣ that is resistant to signal-induced degradation
(I␬B␣M; Van Antwerp et al. 1996). Whereas HeLa cells
transfected with I␬B␣M were sensitized to TNF␣ alone,
the cotransfection of Bfl-1 rescued the cells from cytoly-
sis (Fig. 4c). These results indicated that Bfl-1 can sub-
stitute for NF-␬B to block TNF␣-induced cytolysis.
These data agree with the protective effect of Bfl-1 and
its homolog A1 toward various inducers of cell death
(D’Sa-Eipper et al. 1996; Karsan et al. 1996; Lin et al.
1996). Together, our findings support a role for Bfl-1 as a
survival factor in the NF-␬B-signaling pathway that con-
fers resistance to TNF␣-induced apoptosis.
Immunofluorescence studies localized Bfl-1 to the cy-
toplasm (data not shown). This is compatible with the
subcellular localization of its homolog Bcl-2 (Krajewski
et al. 1993). Consistent with this observation, the inhibi-
tory activity of Bfl-1 toward TNF␣-induced cell death
was similar to that of Bcl-2 in HeLa cells. Whereas Bfl-1
enabled ∼45%–50% of the cells to survive treatment
with TNF␣ plus CHX, Bcl-2 enabled ∼50%–60% of the
cells to escape cytolysis in transient transfection assays
(data not shown). However, further studies are needed to
determine whether Bfl-1 utilizes the same mechanisms
as Bcl-2 to inhibit cell death.
Analysis of the bfl-1 promoter region identified se-
quence elements responsible for its Rel-dependent in-
duction. Nested PCR amplification of adaptor-ligated
human genomic DNA libraries generated products of
∼1.4, 1.3, 0.4, and 0.2 kb (GenomeWalker-kit, Clontech).
All shared a common 3⬘ end derived from the 5⬘ end of
the bfl-1 cDNA and extended 5⬘ into adjacent genomic
sequences. The purified products were directionally
cloned into a promoterless reporter plasmid for analysis
(Fig. 5a). The detailed characterization of the promoter
will be described elsewhere (L.C. Edelstein and C.
Ge´linas, in prep.).
bfl-1 promoter activity was assayed by transient trans-
fection of HeLa cells in the presence or absence of a
CMV-c-rel vector. An IL6␬BCAT reporter plasmid con-
taining three ␬B DNA sites served as a positive control.
As shown in Figure 5b, all four clones showed minimal
basal activity on their own. The cotransfection of
pCMV–c-rel enhanced expression from the −1374/+81
and −1240/+81 promoter constructs by 12- and 9-fold,
respectively. In contrast, the activity of the −367/+81
and −129/+81 constructs was only marginally increased
by c-rel. Our mapping of a consensus NF-␬B DNA site at
position −833 relative to the transcription start site of
bfl-1 agreed with these results (GGGGATTTACC; Fig.
5a). Consistent with these findings, cell treatment with
a physiological inducer of NF-␬B also activated the bfl-1
promoter. As shown in Figure 5c, TNF␣ stimulated CAT
expression from the bfl-1 promoter, similar to its effect
on the control IL6␬BCAT reporter plasmid. Inactivation
of the consensus NF-␬B motif in the context of the bfl-1
promoter region provided direct evidence that bfl-1 is
under Rel/NF-␬B control (GTTT
ATTTACC; −1374/
+81m␬B). Mutation of this NF-␬B site decreased gene
expression significantly in the presence of c-Rel (Fig. 5d).
Figure 3. bfl-1 is highly expressed in human immune tissues
and correlates with endogenous Rel/NF-␬B activity. Multiple
tissue Northern blots (Clontech) human II (lanes 1–5) and hu-
man immune system II (lanes 6–11) were successively hybrid-
ized to
32
P-labeled probes for bfl-1, i
␬
b
␣
,c-rel, relA, and actin.
Figure 4. Bfl-1 suppresses TNF␣-induced cell death. (a) HeLa
cells were cotransfected with pCMV–
␤
-gal, together with an
empty pCMV vector or pCMV–bfl-1. The cells were treated
with CHX alone or together with TNF␣ for 16 hr, stained with
X-gal, and photographed. (b) Quantitation of cell survival upon
expression of bfl-1. The viability of HeLa, HtTA-1, and HT1080
cells transfected as described in a represents the ratio of cells
expressing ␤–gal in wells treated with TNF␣ plus CHX over that
in wells treated with CHX alone. Cells from a minimum of 10
randomly chosen fields were counted. The average survival
from three independent experiments is shown. (c) HeLa cells
were cotransfected with pCMV-
␤
-gal and an empty pCMV vec-
tor or pCMV–I␬B␣M, alone or together with pCMV-bfl-1. The
cells were treated with TNF␣ alone for 16 hr and stained with
X-gal. Cell survival represents the ratio of cells expressing
␤
-gal
in wells treated with TNF␣ over that in wells left untreated.
The average survival from three experiments is shown.
Zong et al.
384 GENES & DEVELOPMENT

Together, these findings demonstrate that bfl-1 is a pro-
survival gene under direct Rel/NF-␬B control.
Prosurvival members of the Bcl-2 family have been
shown to block apoptosis in lymphoid cells under con-
ditions in which NF-␬B activity was inhibited (Wu et al.
1996; for review, see Sonenshein 1997). This raised the
possibility that some members of the Bcl-2 family may
lie downstream of NF-␬B in the survival cascade. Our
demonstration that bfl-1/a1 is a transcriptional target of
NF-␬B provides the first direct evidence that a Bcl-2-fam-
ily member is controlled by NF-␬B proteins. These find-
ings are consistent with previous reports indicating that
bfl-1/a1 gene expression is induced by proinflammatory
cytokines in endothelial, leukemic, and hemopoietic
cells (Moreb and Schweder 1997; Lin et al. 1993; Karsan
et al. 1996b).
Our data also show clearly that bfl-1 can suppress
TNF␣-induced cytolysis. The anti-apoptotic activity of
Bfl-1 in this context agrees with work indicating that
Bfl-1/A1 can confer resistance to a variety of death in-
ducers in different cells (D’Sa-Eipper et al. 1996; Karsan
et al. 1996a; Lin et al. 1996). Thus, Bfl-1
may be viewed as an important player
in the survival pathway. It remains
possible that bfl-1 may act in combina-
tion with other anti-apoptotic genes to
block cell death efficiently in response
to different stimuli and in different
cells. For example, whereas NF-␬Bwas
recently implicated in inducing expres-
sion of the death inhibitors c-IAP1, c-
IAP2, and the TRAF1 (TNFR-associ-
ated factor 1) and TRAF2 factors, all
four proteins must act in combination
to efficiently block TNF-induced apop-
tosis in cells where NF-␬B is inactive
(Chu et al. 1997; Wang et al. 1998; You
et al. 1997). The presence of an NF-␬B
site in the promoter region of the zinc
finger protein A20 suggests that it may
also be under NF-␬B control (Krikos et
al. 1992), although A20 failed to rescue
RelA
−/−
cells from TNF␣-induced cy-
tolysis (Beg and Baltimore 1996). Simi-
larly, the immediate-early response
gene IEX-1L was shown to be involved
in NF-␬B-mediated cell survival, but its
mechanism of action remains to be
clarified (Wu et al. 1998). Preliminary
data from our laboratory agree with
these reports and suggest that other
anti-apoptotic factors are also regulated
by Rel/NF-␬B (C. Chen and C. Ge´linas,
in prep.). It will thus be important to
evaluate how their activities are coor-
dinated by the NF-␬B-signaling path-
way and to determine whether they
function individually or cooperatively
in response to different stimuli and in
different cellular environments.
The coinciding expression of bfl-1 and c-rel in the
white pulp of the spleen, the germinal centers of lym-
phatic tissues, and inflammatory cells (Carrasco et al.
1994; Jung-ha et al. 1998) supports a model whereby Bfl-1
may be a critical factor for carrying out the protective
role of Rel/NF-␬B in the immune system and during the
inflammatory response. Bfl-1 was shown previously to
cooperate with the adenovirus E1A protein in inducing
cell transformation and to be overexpressed in certain
cancers (Choi et al. 1995; D’Sa-Eipper et al. 1996). Al-
though the participation of Bfl-1 in oncogenesis is still a
topic of controversy (Jung-ha et al. 1998), the activation
of Bfl-1 by NF-␬B may also be a means by which NF-␬B
functions in oncogenesis and promotes the resistance of
tumor cells to anti-cancer therapy.
Materials and methods
Cells and endogenous NF-
␬
B activation
Parental HtTA-1 cells and the HtTA-1-derived HtTA–CCR43, HtTA–
RelA, and HtTA–p50 cell clones that expressed c-rel, relA, and p50,re-
spectively, under tetracycline-regulated control have been described
Figure 5. The human bfl-1 promoter contains a consensus NF-␬B site responsible for
its Rel-dependent induction. (a) Schematic representation of CAT reporter gene con-
structs driven by various regions of the human bfl-1 promoter. (b) c-Rel-dependent
transactivation of the bfl-1 promoter. HtTA-1 cells were cotransfected with bfl-1–CAT
reporter plasmids together with pCMV–c-rel or an empty pCMV vector as a control.
The IL6␬BCAT plasmid containing three NF-␬B DNA sites derived from the IL6 pro-
moter was used as a positive control. The average CAT activity from three independent
experiments is shown. (c) TNF␣-inducible activation of the bfl-1 promoter. HtTA-1
cells were transfected with −1374/+81 bfl-1–CAT or the control IL6␬BCAT reporter
plasmid. Where indicated, cells were stimulated with TNF␣ for 6 hr prior to harvest.
The average fold activation from three independent experiments is shown. (d) ␬B site-
dependent activation of the bfl-1 promoter. Cells were cotransfected with wild-type
−1374/+81 bfl-1–CAT or the mutant −1374/+81m␬B–CAT reporter plasmid, with a
mutated NF-␬B site at position −833 (GTTTATTTACC), together with pCMV–c-rel or
an empty CMV vector as a control. IL6␬BCAT was used as a control.
Bfl-1/A1 is under Rel/NF-␬B control
GENES & DEVELOPMENT 385

(Gossen and Bujard 1992; Bash et al. 1997; Zong et al. 1998). Human HeLa
cervical carcinoma cells, HT1080 fibrosarcoma cells and Jurkat T-lym-
phocytic leukemia cells, and mouse 70Z/3 pre-B cells and WEHI-231
mature B-cells were obtained from ATCC. Jurkat–I␬B␣wt and Jurkat–
I␬B␣⌬N T cells were a gift of D.W. Ballard (Chu et al. 1997). Endogenous
NF-␬B activity was induced by treatment with TNF␣ (Sigma; 10 ng/ml)
for 2 hr (WEHI-231) or 3 hr (HT1080), with PMA (50 ng per ml) plus
ionomycin (1 µ
M in DMSO) for 2 hr or DMSO alone as a control (0.05%),
with LPS (10 µg/ml) for 4 hr, or with PMA (100 n
M)for2hr.
Subtractive hybridization and cloning of a bfl-1 cDNA
Poly(A)
+
RNA was isolated from HtTA–CCR43 cells using a QuickPrep
RNA purification kit (Pharmacia). Purified mRNA (2 µg) was reverse
transcribed and subjected to subtractive hybridization with a PCR-Select
cDNA subtraction kit (Clontech). Subtracted cDNA fragments were
cloned in the pCRII vector (Invitrogen) and sequenced (Sequenase 2.0;
U.S. Biochemical). A full-length HA-tagged bfl-1 cDNA clone was ob-
tained by RT–PCR of total RNA from human HeLa cells in two succes-
sive rounds of amplification. The 5⬘ primer used in the first round was
GCGTTCCAGATTACGCTAGCTTGATGACAGACTGTGAATTTG-
GA, and the 3⬘ primer was CTGCTTAAGAGCTCTCAACATGATT-
GCTTCAGG. In the second round, the 5⬘ primer containing an HA
tag was GGATCCGCCATGGCATACCCATATGATGTTCCAGATTA-
CGCT. The 3⬘ bfl-1-specific primer was identical to that used in the first
round. The amplification product was cloned in a pCMV vector for tran-
sient transfection assays (pcDNA3; Invitrogen). The bfl-1 cDNA se-
quence was confirmed with a T7 sequencing kit (Pharmacia).
Northern blot analysis
Total RNA (20 µg) extracted with RNAzol B (TEL-TEST) was fraction-
ated in a 1% agarose–formaldehyde gel and transferred onto a Hybond-
NX membrane (Amersham). The membrane was baked for 10 min at
80°C under vacuum and UV cross-linked with a Stratalinker (Stratagene).
Multiple tissue Northern blots were purchased from Clontech (human,
human II, human immune system II). Probes were generated by random
priming with Klenow polymerase in the presence of [
32
P]dCTP and
[
32
P] dGTP (Feinberg and Vogelstein 1983). Membranes were hybridized
in 5× SSC (0.75
M NaCl, 75 mM Na citrate at pH 7.0), 5× Denhardt’s
solution, 0.5% SDS, and 100 µg/ml sheared salmon sperm DNA at 65°C
overnight. Membranes were washed twice in 2× SSC, 0.1% SDS, and
twice in 1× SSC, 0.1% SDS, at 65°C, followed by autoradiography.
TNF
␣
-induced apoptosis
Cell resistance to TNF␣-induced apoptosis was assayed as described
(White et al. 1992). Cells (3 × 10
6
) were incubated with pCMV–
␤
-gal
(5µg), together with pCMV–bfl-1 (15 µg) or an empty pCMV vector as a
control, and electroporated at 220 V, 960 µF using a Bio-Rad Gene Pulser.
The cells were then distributed equally into two 35-mm wells. After 24
hr, the cells were treated with CHX alone (30 µg/ml) or together with
TNF␣ (10 ng/ml) for 16 hr. The cells were fixed and stained with X-gal
and photographed at a magnification of 200x. In assays of cell death
performed in the absence of CHX, HeLa cells were coelectroporated with
pCMV–
␤
-gal (3 µg), an empty CMV vector, or pCMV–I␬B␣M (12 µg) to
constitutively repress NF-␬B, alone or together with pCMV–bfl-1 (6 µg).
Cells were then distributed equally into two 35-mm wells and treated 24
hr later with TNF␣ (10 ng/ml) for 16 hr.
Cloning of the human bfl-1 promoter, transient CAT assays,
and mutagenesis
The human bfl-1 promoter region was isolated by nested PCR amplifi-
cation with a GenomeWalker-PromoterFinder kit (Clontech) and cloned
in a promoterless vector expressing a CAT reporter gene (pCAT-basic;
Promega). bfl-1 promoter activity was analyzed by transient transfection
of HtTA-1 cells with bfl-1–CAT reporter plasmids (3 µg) in the presence
of a CMV–c-rel expression vector (1 µg; Xu et al. 1993) or an empty pCMV
vector as a control. Where indicated, cells were treated with TNF␣ (10
ng/ml) for 6 hr before harvest. An IL6␬BCAT reporter plasmid containing
three NF-␬B DNA sites from the IL6 promoter was used as a positive
control (Xu et al. 1993). The −1374/+81 bfl-1 promoter region cloned in
pAlter-1 was subjected to site-directed mutagenesis to inactivate the
consensus NF-␬B motif at position −833 (GTTT
ATTTACC, −1374/
+81m␬B; Altered Sites Mutagenesis System, Promega). Mutation of the
consensus NF-␬B site was confirmed by sequencing.
Acknowledgments
We are very grateful to C. Labrie for allowing C.C. to clone HA-bfl-1 in
his laboratory, to D.W. Ballard for Jurkat–I␬B␣wt and Jurkat–I␬B␣⌬N
cells, and to H. Bujard for the gift of HtTA-1 cells. We thank A. Rabson,
B. Rayet, A. Shatkin, and E. White for helpful comments on the manu-
script. This work was supported by grants from the National Institutes of
Health (NIH CA54999), The Council for Tobacco Research USA (4175),
and by the New Jersey Commission on Science and Technology. L.C.E. is
supported by NIH Biotechnology pre-doctoral training grant GM08339.
C.C. is a postdoctoral fellow of the New Jersey Commission on Cancer
Research and The Foundation of UMDNJ.
The publication costs of this article were defrayed in part by payment
of page charges. This article must therefore be hereby marked ‘advertise-
ment’ in accordance with 18 USC section 1734 solely to indicate this
fact.
References
Adams, J.M. and S. Cory. 1998. The Bcl-2 protein family: Arbiters of cell
survival. Science 281:1322–1326.
Bash, J., W.-X. Zong, and C. Ge´linas. 1997. c-Rel arrests the proliferation
of HeLa cells and affects critical regulators of the G1/S-phase tran-
sition. Mol. Cell. Biol. 17: 6526–6536.
Beg, A.A. and D. Baltimore. 1996. An essential role for NF-␬B in prevent-
ing TNF␣-induced cell death. Science 274: 782–784.
Beg, A.A., W.C. Sha, R.T. Bronson, and D. Baltimore. 1995. Constitutive
NF-␬B activation, enhanced granulopoiesis, and neonatal lethality in
I␬B␣-deficient mice. Genes & Dev. 9: 2736–2746.
Cai, Z., M. Korner, N. Tarantino, and S. Chouaib. 1997. I␬B␣ overexpres-
sion in human breast carcinoma MCF7 cells inhibits nuclear factor-
␬B activation but not tumor necrosis factor ␣-induced apoptosis. J.
Biol. Chem. 272: 96–101.
Carrasco, D., F. Weih, and R. Bravo. 1994. Developmental expression of
the mouse c-rel proto-oncogene in hematopoietic organs. Develop-
ment 120: 2991–3004.
Choi, S.S., I.-C. Park, J.W. Yun, Y.C. Sung, S.I. Hong, and H.-S. Shin.
1995. A novel Bcl-2 related gene, Bfl-1, is overexpressed in stomach
cancer and preferentially expressed in bone marrow. Oncogene
11: 1693–1698.
Chu, Z.-L., T.A. McKinsey, L. Liu, J.J. Gentry, M.H. Malim, and D.W.
Ballard. 1997. Suppression of tumor necrosis factor-induced cell
death by inhibitor of apoptosis c-IAP2 is under NF-␬B control. Proc.
Natl. Acad. Sci. 94: 10057–10062.
D’Sa-Eipper, C., T. Subramanian, and G. Chinnadurai. 1996. Bfl-1, a Bcl-2
homologue, suppresses p53-induced apoptosis and exhibits potent co-
operative transforming activity. Cancer Res. 56: 3879–3882.
Feinberg, A.P. and B. Vogelstein. 1983. A technique for radiolabeling
DNA restriction endonuclease fragments to high specific activity.
Anal. Biochem. 132: 6–13.
Gossen, M. and H. Bujard. 1992. Tight control of gene expression in
mammalian cells by tetracycline-responsive promoters. Proc. Natl.
Acad. Sci. 89: 5547–5551.
Jung-ha, H., D. Kim, S.-B. Lee, S.-I. Hone, S.-Y. Park, J. Huh, C.-W. Kim,
S.-S. Kim, Y. Lee, S.S. Choi, and H.-S. Shin. 1998. Expression of Bfl-1
in normal and tumor tissues: Bfl-1 overexpression in cancer is attrib-
utable to its preferential expression in infiltrating inflammatory
cells. Hum. Pathol. 29: 723–728.
Karsan, A., E. Yee, and J.M. Harlan. 1996a. Endothelial cell death induced
by tumor necrosis factor ␣ is inhibited by the Bcl-2 family member,
A1. J. Biol. Chem. 271: 27201–27204.
Karsan, A., E. Yee, K. Kaushansky, and J.M. Harlan. 1996b. Cloning of a
human Bcl-2 homologue: Inflammatory cytokines induce human A1
in cultured endothelial cells. Blood 87: 3089–3096.
Krajewski, S., S. Tanaka, S. Takayama, M.J. Schibler, W. Fenton, and J.C.
Reed. 1993. Investigation of the subcellular distribution of the bcl-2
oncoprotein: Residence in the nuclear envelope, endoplasmic reticu-
lum, and outer mitochondrial membranes. Cancer Res. 53: 4701–
4714.
Krikos, A., C.D. Laherty, and V.M. Dixit. 1992. Transcriptional activa-
tion of the tumor necrosis factor ␣-inducible zinc finger protein, A20,
is mediated by ␬B elements. J. Biol. Chem. 267: 17971–17976.
Lin, E.Y., A. Orlofsky, M.S. Berger, and M.B. Prystowsky. 1993. Charac-
Zong et al.
386 GENES & DEVELOPMENT

terization of A1, a novel hemopoietic-specific early-response gene
with sequence similarity to bcl-2. J. Immunol. 151: 1979–1988.
Lin, E.Y., A. Orlofsky, H.G. Wang, J.C. Reed, and M.B. Prystowsky. 1996.
A1, a Bcl-2 family member, prolongs cell survival and permits my-
eloid differentiation. Blood 87: 983–992.
Liou, H.C., W.C. Sha, M.L. Scott, and D. Baltimore. 1994. Sequential
induction of NF-␬B/Rel family proteins during B-cell terminal differ-
entiation. Mol. Cell. Biol. 14: 5349–5359.
Liu, Z., H. Hsu, D.V. Goeddel, and M. Karin. 1996. Dissection of TNF
receptor 1 effector functions: JNK activation is not linked to apopto-
sis while NF-␬B activation prevents cell death. Cell 87: 565–576.
Miyamoto, S., M.J. Schmitt, and I.M. Verma. 1994. Qualitative changes
in the subunit composition of ␬B-binding complexes during murine
B-cell differentiation. Proc. Natl. Acad. Sci. 91: 5056–5060.
Moreb, J.S. and M. Schweder. 1997. Human A1, a Bcl-2-related gene, is
induced in leukemic cells by cytokines as well as differentiating fac-
tors. Leukemia 11: 998–1004.
Nagata, S. 1997. Apoptosis by death factor. Cell 88: 355–365.
Reed, J.C. 1996. Mechanisms of Bcl-2 family protein function and dys-
function in health and disease. Behring Inst. Mitt. 97: 72–100.
Sonenshein, G.E. 1997. Rel/NF-␬B transcription factors and the control
of apoptosis. Sem. Cancer Biol. 8: 113–119.
Van Antwerp, D.J., S.J. Martin, T. Kafri, D.R. Green, and I.M. Verma.
1996. Suppression of TNF␣-induced apoptosis by NF-␬B. Science
274: 787–789.
Van Antwerp, D.J., S.J. Martin, I.M. Verma, and D.R. Green. 1998. Inhi-
bition of TNF␣-induced apoptosis by NF-␬B. Trends Cell Biol. 8 107–
111.
Wang, C.-Y., M.W. Mayo, R.G. Korneluk, D.V. Goeddel, and A.S. Bald-
win, Jr. 1998. NF-␬B antiapoptosis: induction of TRAF1 and TRAF2
and c-IAP1 and c-IAP2 to suppress caspase-8 activation. Science
281: 1680–1683.
Wang, C.-Y., M.W. Mayo, and A.S. Baldwin, Jr. 1996. TNF␣ and cancer
therapy-induced apoptosis: potentiation by inhibition of NF-␬B. Sci-
ence 274: 784–787.
White, D.W., A. Roy, and T.D. Gilmore. 1995. The v-Rel oncoprotein
blocks apoptosis and proteolysis of I␬B␣ in transformed chicken
spleen cells. Oncogene 10: 857–868.
White, E., P. Sabbatini, M. Debbas, W.S.M. Wold, D.I. Kusher, and L.R.
Gooding. 1992. The 19-kilodalton adenovirus E1B transforming pro-
tein inhibits programmed cell death and prevents cytolysis by tumor
necrosis factor ␣. Mol. Cell. Biol. 12: 2570–2580.
Wu, M., H. Lee, R.E. Bellas, S.L. Schauer, M. Arsura, D. Katz, M.J.
FitzGerald, T.L. Rothstein, D.H. Sherr, and G.E. Sonenshein. 1996.
Inhibition of NF-␬B/Rel induces apoptosis of murine B cells. EMBO
J. 15: 4682–4690.
Wu, M.X., Z. Ao, K.V.S. Prasad, R. Wu, and S.F. Schlossman. 1998. IEX-
1L, an apoptosis inhibitor involved in NF-␬B-mediated cell survival.
Science 281: 998–1001.
Xu, X., C. Prorock, H. Ishikawa, E. Maldonado, Y. Ito, and C. Ge´linas.
1993. Functional interaction of the v-Rel and c-Rel oncoproteins with
the TATA-binding protein and association with transcription factor
IIB. Mol. Cell. Biol. 13: 6733–6741.
You, M., P.-T. Ku, R. Hrdlickova, and H.R. Bose. 1997. ch-IAP1, a mem-
ber of the inhibitor-of-apoptosis protein family, is a mediator of the
antiapoptotic activity of the v-Rel oncoprotein. Mol. Cell. Biol.
17: 7328–7341.
Zong, W.-X., M. Farrell, J. Bash, and C. Ge´linas. 1997. v-Rel prevents
apoptosis in transformed lymphoid cells and blocks TNF␣-induced
cell death. Oncogene 15: 971–980.
Zong, W.-X., J. Bash, and C. Ge´linas. 1998. Rel blocks both anti-Fas- and
TNF␣-induced apoptosis and an intact Rel transactivation domain is
essential for this effect. Cell Death Differ. 5: 963–972.
Bfl-1/A1 is under Rel/NF-␬B control
GENES & DEVELOPMENT 387